Lock-up clutches

[lecroy]

I have not seen the multi-stage up close. Just pictures on their site. Looking at the top two, it looks like the two are the same. I just assumed the only difference was one was shipped without the springs.

I agree with you. A logger is the only way to go. I saw that one MPS has one with just the two speed inputs for under $600.

http://www.mpsracing.com/products/RPM/datalogger.asp

This is about all you would need to set one up.

I'm sticking with my home made logger from hell. I need to shrink it down and make it open design. Without the sensors, I think I could make a kit available for under $100. Let people do their own thing with it.

http://dickey.intell-a-sys.com/logger3.html

[rosco15]

I would love a logger that just did rpm. Really see what the clutch is doing. Would make tuning really easy. The mps unit is way too rich for my blood.

Could you set up your data logger to do rpm using the tach output from the schnitz box. You can set the output to either 1 or 2 output signals per revolution.

[lecroy]

I would love a logger that just did rpm. Really see what the clutch is doing. Would make tuning really easy. The mps unit is way too rich for my blood.

My first goal of this was to make something that people could use as a real tool and change it to fit their needs. Not like all the $1000 - $10000
systems I was looking at, most with 10Hz sample rates and 8 bits of resolution. My second goal was to make it so most people could afford it.
Sort of like the whole MegaSquirt EFI project. The nice thing is that the logger is VERY VERY simple compared to an EFI. I think the program that runs it is maybe 50 lines of C code right now. Because I used all open sourced tools to develop the unit, anyone can download them and make changes.

There was not much interest in it. So for now I am spending time making my prototype better.

Could you set up your data logger to do rpm using the tach output from the schnitz box. You can set the output to either 1 or 2 output signals per revolution.

It seems that if the ignition guys did one thing right, it was the tach signal. The Dyna SP 4000, MC-2 and my MC-4 all work the same. Because they need to work with other equipment like RPM switches, tachometers, shift lights and such they all seem to follow a standard. So, yes it will work.

To give you an idea of what you could see using a simple two channel logger, this first graph is the clutch slippage in percent. The logger knows the crank speed from the tach and measures the output shaft speed using a sensor. You enter all of your data about your engine and the program then knows what the ratio of the two should be. If the engine is spining at 1000 RPM and the output shaft is not moving, it's a good indication that the clutch is not engaged, or zero percent.

Two very interesting things about this graph. From the time I released the button to the time the clutch starts to engage is 0.6 seconds. Then it takes another 0.5 seconds to fully engage. But even more strange is the funny shape. It's not smooth like you would expect. And you thought my one video clip was soft!! LOL This was me playing with a restrictor on the output of the air cylinder to slow the release to try and keep the nose up. It would be like you letting your clutch lever out slow. The air ram chatters when I do this causing the oscillations. Even though I could tune for the delay, it was not consistant at all and caused the steel plates to get hot spots.


So I dropped this idea and started experimenting with lower static pressures. So, same bike, engine, tire, etc. No restrictor and very low spring force with a lot of arm mass.

Notice now that when I release the button, the clutch starts to engage after 0.05 seconds and is fully engaged after 0.22 seconds. More important is the shape.


The clutch gets to about 30% (slope A) and stays there for about 0.07 seconds doing a nice slip until the massive weights start to kick in (slope B). This keeps the shock down, tire from spinning and I would suspect allow the bike to run at Norwalk, or in my driveway with about the same results. This is the data from the last driveway video clip BTW. This seems to also solve the hot spot problem.

The next step would be to start pushing slope B back to slope A to get the bike to react faster. The only problem is if you push things too far you get something like this.

This is a graph of the output shaft in MPH. Now my engine tuning is not good enough to ever win me an award, but it sure is not so bad that it's causing this. So softer is slower, but it sure beats a 10" 4.5PSI bouncing ball. But then again, a little better driveway prep would go a long way to help this out. I need to buy some VHT for home use.

Ok, one last thing before closing the book on this subject. So, I can plot two shaft speeds in MPH, RPM, frequency, or take a ratio of the two there is one other interesting thing. This last graph shows how the bike accelerates. This graph is showing the change in output shaft RPM every 0.01 seconds. Anything greater than zero and the bike is speeding up, anything negative and the bike is slowing down.

You can see how it starts out nice and smooth then things start to get a little funny. At 5.2 seconds into the run the bike actually starts to slow down. Notice the bottom graph is showing boost pressure in PSIG. Notice that every time the bike shifts, the boost pressure rolls off. My bike uses a stock transmission so I kill on every shift for 0.05 seconds. Notice that the negative spike in the first graph happens after the bike shifts. The tire spun. We could see that the tire was not holding the track very well anyway prior to it giving way. Of course, higher pressures and we are making a tad bit more power.

Ok, the question should be how do I know the tire slipped and not the clutch. Well this is the whole point of the logger. I can plot the clutch slippage as before and see. If the clutch is not slipping, then it must be the tire.

Well, hope it at least gives you an idea of what you could do with two speed sensors and a logger if you did not know already. I get a lot of laughs with all of the electronics on my old bike.

[marcovandevelde]

Hi Lecroy,
Sorry, I never considered my amateur imput to be of value to you.

It was sold to me as a 2-stage lockup. I completely mucked up the 2 stage as I had to put max weight on all arms ofcourse. It is not meant to be fitted with springs as std., but I see how you could do that. Needs some experimenting, but as I see you are the ''science-boi'' here this couldn't be a problem for you, could it?

The spacer is at the welder at the mo to have a neat oilreturn made into it, when it's back I'll measure the width for you.

Greetz, Marco.

[lecroy]

Hi Lecroy,
Sorry, I never considered my amateur imput to be of value to you.

I am sorry if I wrote something to make you feel this way. I consider myself an amateur at best. I would consider myself the waterboy on a football team, except I get to play with the ball once in a while. Helping out as a pit guy has really taught me a lot about the sport. I even pick good pointers from the kid. He is a natural. If we could just get him interested in setting up his own bikes.

It was sold to me as a 2-stage lockup. I completely mucked up the 2 stage as I had to put max weight on all arms ofcourse. It is not meant to be fitted with springs as std., but I see how you could do that. Needs some experimenting, but as I see you are the ''science-boi'' here this couldn't be a problem for you, could it?

It would be interesting to see the two (multi and 2 stage units). You could add more weight by the looks of it. On my bike, this can cause problems as well. Everytime I play with a clutch it seems to be a balancing act to get it right. I would love to spend a weekend at a real race on a T/F team just to see how they tune them.

I wanted to add a return line to the new cover and just use the stock oil cap but there was just no room. The oil cap works good and does not look too bad on the bike so I plan to just leave it.

[gsbill]

We ran the mtc lockup on our nitrous bike. Only to help with clutch feel and to keep the clutch from slipping with a sticky race tire. Our bike made about 250hp. I rode it on the street too. More than likely you dont need a lock up on the street. You will never get the traction to get any benefit from it. You will spin and smoke the rear tire first.

[marcovandevelde]

Hi lecroy,

I am sorry if I wrote something to make you feel this way.

No, you didn't do anything wrong! Only thing I meant to say is that I only drive my bike on the streets and don't race it. This way things are much less critical than in your application and therefore I was surprised to find interest.

Anyway, you wanted to know the width of the MTC spacer: 33mm.

It's a good quality cast that lends itself to be fitted with a coupler for the return-oil from the turbo.

Marco.

[lecroy]

I have been looking for a low cost solution for a data logger for a while now and had started to make an open source unit. Now I think I may have a better idea. The model airplane guys also use data loggers and there is a company in the USA that happens to make one that appears to be a very good. Check out their website:

http://www.eagletreesystems.com/

They offer a few different units, all at a fraction of the cost of what the lowest cost logger I could find. They are also very light weight. All of their units have two 16-bit RPM inputs. This is just what you would need to do the clutch measurements I was showing. They can also measure EGT, pressure and a host of other things. The analog inputs are 12-bit for the low cost systems, which is more than enough for most applications. And again, they are very low cost! I'm talking $150. I am not sure I could sell my kit for this much.



I spoke with them about using their products for motorcycle drag racing and it sounds like they would be interested in supporting it. They would be willing to make changes to their software and hardware to fit our needs. The only question is if there is enough interest.


Please take the time to contact them expressing any interest you may have.

Info@EagleTreeSystems.com

Mark

DataX